1. Meet in Map:
We will use algorithms to filter out nearby users and display them on the map. Our recommendation algorithm utilizes deep learning to extract personalized features and latent variables, effectively analyzing and integrating conventional information such as geographic location, user interests, occupational fields, and the groups they belong to. Additionally, it can utilize large language models to extract language styles and topics of interest based on personal chat corpus. Ultimately, multiple distance functions are used to rank the high-dimensional feature latent variables for optimal recommendations.
2. Meet in Group:
Our product promotes online growth through a "Group by Group" and "Club by Club" approach, aiming to ensure a sufficiently dense user population within a small area while maintaining mutual trust and basic understanding among users. This approach allows club members and high-quality community members the opportunity to meet offline and engage in face-to-face conversations.
3. Meet up:
Our primary goal is to facilitate genuine face-to-face meetups among users. Our algorithms and modes ensure the trustworthiness and immediacy of meetups. The average number of meetups per user per week is a key performance indicator (KPI) that we prioritize in our product.